bk-2009-1015.ch012

Downloaded by STANFORD UNIV GREEN LIBR on June 26, 2012 | http://pubs.acs.org Publication Date (Web): December 19, 2009 | doi: 10.1021/bk-2009-1015.ch012

Chapter 12

Exposure of Adults and Children to Organophosphorus Insecticides used in Flea Collars on Pet Dogs Janice E. Chambers1 and M. Keith Davis1 1

Department of Basic Sciences/Center for Environmental Health Sciences, College of Veterinary Medicine, Mississippi State University, Mississippi State, MS 39762

Our laboratories have investigated the exposure of people to the organophosphorus insecticides chlorpyrifos and tetrachlorvinphos that were contained in flea collars used on their pet dogs. Long-term studies conducted over much of the recommended useful lifetime of the collars indicated that residues of both insecticides were transferable to white cotton gloves which were used to rub the fur of the dogs, with the tetrachlorvinphos residues considerably higher than the chlorpyrifos residues. In short-term studies, residues of both insecticides were transferred to tee shirts worn by the children. In these same studies the urinary metabolite of chlorpyrifos was perhaps slightly elevated over the background level in children but was not elevated over the background level in adults. In contrast the urinary metabolite of tetrachlorvinphos was substantially elevated over background levels in both children and adults. However, the significance of these findings to risk assessment is not known.

© 2009 American Chemical Society

163

In Pesticides in Household, Structural and Residential Pest Management; Peterson, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2010.


164 A large proportion of homes in the United States have pet dogs, roughly about 40%, and of those homes, about two thirds have children (1). Since ectoparasites are both a medical problem and a nuisance to pets and their owners, many pet owners use flea and tick control products. One type of flea and tick control product that is common, easy to obtain and available without a prescription (and therefore is relatively inexpensive) is a plastic collar embedded with an insecticide. Organophosphorus insecticides have been the insecticides of choice for a number of these collars. Organophosphorus insecticides or their active metabolites (i.e., oxons) are anticholinesterases; the insecticides or their oxons phosphorylate the active site serine of the acetylcholinesterase, a reaction that inhibits the catalytic action of the enzyme allowing the accumulation of the neurotransmitter acetylcholine in synapses and neuromuscular junctions (2). At sufficiently high levels of acetylcholinesterase inhibition, a variety of autonomic nervous system effects occur as can tremors and convulsions, with death occurring from respiratory failure at lethal levels of exposure. The organophosphorus insecticides have been subject to the Food Quality Protection Act-mandated cumulative risk assessment as these insecticides act toxicologically through a common mechanism of toxicity, i.e., the inhibition of acetylcholinesterase. As routes of exposure are considered for a cumulative risk assessment, certainly any sources of insecticide from residential uses would need to be considered. The use of insecticides for flea and tick control on pets would be a potential source of insecticide exposure in residential settings, and this potential exposure has formed the basis for the study of several flea and tick control products in our laboratories. Little is known about how much insecticide might transfer to people in contact with their pets from any topically-applied insecticide (such as from dips or shampoos) or any insecticide from a product in which the insecticide diffuses out of the product and deposits on the fur (such as from collars). In the case of some of these products, the concentration of insecticide is quite high either directly deposited on the animal’s fur or embedded in the product, and could provide a source of appreciable exposure to the people in contact with their pets. Information on this potential source of exposure is necessary in order to determine whether this exposure is of a magnitude of concern, and, if so, for this source to be included in the cumulative risk assessments. Protocols for the standard measurement of transferable residues from such ectoparasite treatments were not established when our group started these studies, so we developed protocols involving petting the treated dogs with new laundered and solvent-extracted white cotton gloves for a set 5-minute period in a prescribed area of the dog’s fur. The gloves were subsequently extracted with appropriate solvents, and the residues were quantified by analytical chemistry methods. These data were considered to represent an estimate of the level of residues that were available for transfer from the fur of the dog to a person. Our initial studies were on two over-the-counter dips containing either chlorpyrifos or phosmet (3, 4). Subsequently we studied two collars that contained either chlorpyrifos or tetrachlorvinphos (5, 6). Concurrently with these latter two studies of collars, we were also able to do biomonitoring of urinary metabolites of the insecticide in an adult and in a child in each household participating in the



165 study. For all of these studies the Institutional Animal Care and Use Committee (IACUC) of Mississippi State University provided prior approval of the protocols used for the handling of the dogs, and the Institutional Review Board (IRB) for Research involving Human Subjects of Mississippi State University provided prior approval of the protocols used for the recruitment and tests involving human subjects. The informed consent of the adult participants and the assent of the children were obtained prior to their participation. The protocol we developed to quantify transferable residues from the dog’s fur was to rub the dog in a back-and-forth manner for a 5-minute period in a set measured area of the dog’s fur. These gloves were light-weight white cotton gloves, not previously used, which were laundered and solvent-extracted prior to use. The sampling occurred at three locations on the dog: on a place on the back near the tail, distant from the collar; on the neck with the collar removed; and on the neck with the collar in place (over the collar). The gloves were then placed into glass bottles which had been pre-rinsed with solvent. Standard solvent extractions occurred, and the insecticide was quantified using gas chromatography with electron capture detection. (We wish to note that the dogs seemed to be very willing participants for these sampling sessions). For the studies where we conducted biomonitoring, in the participating household an adult (either sex) and a child (either sex) between the ages of 3 and 12 years provided first morning void urine samples. These samples were acidhydrolyzed and extracted, and the insecticide metabolite was quantified using gas chromatography with electron capture detection (chlorpyrifos metabolite) or gas chromatography with mass spectrometry detection (tetrachlorvinphos metabolite). In these studies the participating child also wore a new laundered white cotton tee shirt for several hours on the afternoon prior to the collection of the urine sample. A section was cut from the front of the tee shirt and, similar to the procedure with the gloves, the fabric was extracted with appropriate solvents, and the insecticide was quantified using gas chromatography with electron capture detection. The rationale for the tee shirt studies was that the tee shirt might be a suitable surrogate for the potential transferable residues and therefore for the level of exposure. Statistical analysis of the data tested for correlations between the residues observed and certain characteristics of the dog such as fur length or behavioral characteristics such as the length of time the child spent with the pet dog.

Chlorpyrifos Chlorpyrifos is a very widely used insecticide, and it displays a moderate level of acute oral mammalian toxicity but a low level of acute dermal toxicity (rabbit dermal LD50 for chlorpyrifos of 2 g/Kg) (7). While its uses have been curtailed in recent years, it has been so widely used in the past that most of the United States population seems to have exposure to chlorpyrifos, as evidenced by the presence of the chlorpyrifos metabolite trichloropyridinol (TCP) in the majority of the American population sampled in such biomonitoring studies as the National Health and Nutrition Examination Survey (NHANES) (8).



166 We performed two studies using an over-the-counter chlorpyrifoscontaining collar which contained 8% chlorpyrifos (5). The long-term study had sampling of transferable residues for 168 days, which was a substantial fraction of time of the 11 months for which the collar was recommended. The short-term study selected a week from the data of the long term study (week 3 after collar placement) during which peak transferable residues occurred. Transferable residues from the dog’s fur were obtained on two occasions during this week. Urine samples were obtained from an adult and a child in the household of the dog on 5 consecutive days with the child wearing the white cotton tee shirts on 5 consecutive days using the days preceding the urinary metabolite collection. Pretreatment residues on gloves (i.e., background levels) for the five minute rubbing period were measurable, but low, between 1 and 2 µg/glove. Transferable residues increased over the first few days following collar placement, and then remained relatively constant over the rest of the 168 day sampling period. As expected, transferable residues obtained directly over the collar were the highest (about 170 to almost 400 µg/glove), followed by residues from the neck without the collar in place (about 50 to 250 µg/glove), and lowest from the back distant from the collar (from 2 to 14 µg/glove (Table I). The data indicated that relatively low levels of insecticide migrated into the fur a long distance from the collar. The same patterns were apparent in the short term study which obtained transferable residues from fur on days 14 and 20 following collar placement: neck with collar in place, 430-500 µg/glove; neck with collar removed, 280-350 µg/glove; and back, 7-10 µg/glove (Table II). In the short-term study, about 120-200 ng chlorpyrifos/g shirt was observed (Table III). It was originally thought that the tee shirt might be a useful surrogate for biomonitoring data, since obtaining urine samples is somewhat problematical in children; however, the data from the tee shirts did not correlate well with the urinary metabolite data or with any of the activity records provided by the parents on length of time the child spent with the pet dog. Table I. Chlorpyrifos Concentrations on Cotton Gloves for the Long-term Study (concentration ± standard error) Day 0 (4 Hour) 1 3 7 14 28 56 84 112 140 168

Back (µg/glove) 2.17 ± 1.01 4.86 ± 1.90 5.73 ± 1.00 4.11 ± 0.59 6.30 ± 1.51 12.65 ± 3.81 8.45 ± 1.98 14.18 ± 3.60 10.15 ± 1.77 10.01 ± 2.43 9.16 ± 2.50

Neck (µg/glove) 49 ± 8 128 ± 35 125 ± 19 118 ± 15 242 ± 34 241 ± 38 238 ± 41 216 ± 26 252 ± 38 220 ± 26 194 ± 22

Collar (µg/glove) 168 ± 27 190 ± 29 218 ± 25 184 ± 21 391 ± 75 318 ± 27 350 ± 44 310 ± 40 387 ± 49 377 ± 61 313 ± 34

NOTE: Pretreatment value was 1.74 ± 0.49 µg/glove.


167 Table II. Chlorpyrifos Residues on Cotton Gloves for the Short-term Studies (concentration ± standard error) Day Chlorpyrifos

Back (µg/glove)

Neck (µg/glove)

Collar (µg/glove)

14

10.14 ± 2.31

356 ± 66

503 ± 86

20

6.53 ± 1.73

279 ± 60

434 ± 80


NOTE: Pretreatment value for chlorpyrifos was 1.43 ± 0.77 µg/glove.

Table III. Chlorpyrifos Tee Shirt Residues for the Short-term Studies (concentration ± standard error)

Chlorpyrifos

Day 15 16 17 18 19

ng/g Shirt 134.06 ± 66.03 201.46 ± 67.15 126.12 ± 25.81 133.89 ± 45.49 172.03 ± 68.96

Even though chlorpyrifos uses have been reduced in recent years, it is still a widely used insecticide and many, if not most, people have exposure to it, and its metabolite, TCP, is prevalent in human urine. For this reason, urine samples taken from the subjects, both adults and children, had non-zero levels of TCP, with averages across both the long term and short term studies of 8.0-9.2 ng/ml in adults and 10.5-13.5 ng/ml in children (Tables IV and V). In the long term (168 day) study, urinary TCP levels averaged 8.7-10.9 ng/ml for adults and 11.7-16.0 in children, with samples taken on 5 occasions during the 168 day test (Table IV). During the short term study, we concentrated on daily samples for 5 days to assess day-to-day variation within individuals; samples were taken during the third week following placement of the collar on the pet dog. Urinary TCP levels were from 6.9-9.9 ng/ml in adults and from 11.2-15.9 ng/ml in children (Table V). In both of these studies the levels of TCP were higher in the urine from children than in the urine from adults. None of the post-treatment values were significantly different (P < 0.05) from the pretreatment values in either the long term or the short term studies in either adults or children. In both studies, the post-treatment adult TCP levels were both above and below the pretreatment levels, so it appears that no enhanced exposure of adults occurred from the flea collar on the pet dog. However, in the long term study, the posttreatment children’s urinary TCP levels were generally higher than the pretreatment value, even though not statistically significant, while in the short term study, the post-treatment values were both above and below the pretreatment values. Therefore, it is difficult to conclude whether there is enhanced exposure



168 of children to chlorpyrifos from this flea collar because one study suggested there was, but the other study suggested there was not. Certainly it is logical to think that children would in all likelihood be in greater contact with a pet dog than the adults of the household, so there is concern regarding this as a potential exposure route, but the numbers are not convincing that this is the case. The fact that the pretreatment TCP level in children in the long term study (10.5 ng/ml) was lower than the pretreatment value in children in the short term study (13.5 ng/ml) suggests the possibility that the apparent greater levels in children in the long term study are an artifact of the lower pretreatment number which was used for the comparison. The pretreatment level in children for the short term study (13.5 ng/ml) is in the middle of the range of the long term study post-treatment values. Therefore, we can conclude that adults do not seem to receive an enhanced exposure to chlorpyrifos from this collar. However, from these two studies we remain unable to conclude definitively whether children received an enhanced exposure to chlorpyrifos from this collar; however, if they did, it was only a very small amount and did not raise the level of exposure appreciably over the background levels. Table IV. Urinary TCP Concentrations for the Long-term Study (concentration ± standard error) TCP (ng/mL Urine)

Day 3 7 28 84 168

Adult 8.75 ± 1.35 9.43 ± 2.09 9.80 ± 1.42 10.88 ± 1.82 8.72 ± 1.34

Child 11.70 ± 2.21 13.01 ± 2.04 16.01 ± 2.53 15.10 ± 1.95 12.08 ± 2.12

NOTE: Pretreatment values for TCP = 9.15 ± 1.62 for Adults and 10.49 ± 1.83 for children.

Table V. Urinary TCP Concentrations for the Short-term Studies (concentration ± standard error) TCP (ng/mL Urine)

Day 16 17 18 19 20

Adult 6.92 ± 1.31 7.12 ± 1.49 9.05 ± 1.95 9.92 ± 2.18 9.68 ± 1.61

Child 14.79 ± 2.88 13.57 ± 2.47 15.31 ± 2.39 15.92 ± 2.23 11.15 ± 1.18

NOTE: Pretreatment values for TCP = 7.95 ± 0.86 for adults and 13.54 ± 1.62 for children.


169


Tetrachlorvinphos Another organophosphorus insecticide that has been used routinely in flea collars is tetrachlorvinphos. Tetrachlorvinphos (TCVP) is a low toxicity pesticide (rabbit dermal LD50 of 2.5g/Kg) (9). The TCVP collar was also an over-the-counter collar and contained 14% TCVP. It had a recommended use time of 4 months, so we conducted a long term study over 112 days (6). Similar to the protocol we described above, transferable residue samples were obtained by rubbing the dog’s fur for 5 minute sampling periods over the same three regions of the dog: the back distant from the collar, the neck with the collar removed, and the neck with the collar in place. As expected, and similar to the results described above for chlorpyrifos, the lowest residues were obtained on the back distant from the collar and the highest residues were obtained by rubbing over the collar. In contrast to the chlorpyrifos collar, there was a peak of transferable residues obtained over the first two weeks after collar placement with considerably lower levels of residues after the first two weeks. Also in contrast to chlorpyrifos, the levels of transferable residues were considerably higher by about 2 orders of magnitude (Table VI). Table VI. Tetrachlorvinphos Concentrations on Cotton Gloves for the Long Term Study (concentration ± standard error) Day 0 (4 Hour) 3 7 14 28 56 84

Back (µg/glove) 185 ± 26 261 ± 52 177 ± 27 152 ± 22 144 ± 15 80 ± 18 36 ± 8

Neck (µg/glove) 3,530 ± 564 8,042 ± 706 8,674 ± 860 6,062 ± 902 3,844 ± 597 2,802 ± 635 953 ± 168

Collar (µg/glove) 14,340 ± 1,531 23,728 ± 2,125 24,039 ± 3,972 19,309 ± 3,252 12,568 ± 2,086 12,426 ± 2,362 4,956 ± 1,049

112

34 ± 8

549 ± 148

3,267 ± 982

NOTE: Pretreatment value was 0.22 ± 0.15 µg/glove.

A second study was also conducted with this collar, and, similar to above, was a short term study and collected tee shirt and biomonitoring data in addition to the transferable residues (6). This study was conducted over the second week after collar placement during the period that was identified in the long term study as the time of peak transferable residues. Transferable residues from the dog’s fur were similar to those obtained in the long term study (Table VII). Tee shirt residues obtained from the front of the shirt which the child wore on the day before sampling for the urinary metabolite 2,4,5-trichloromandelic acid (TCMA) were in the range of about 1,000-2,000 ng/g shirt (Table VIII). These residues were about an order of magnitude greater than those obtained from the chlorpyrifos collar.


170 Table VII. Tetrachlorvinphos Residues on Cotton Gloves for the Shortterm Studies (concentration ± standard error) Day

Back (µg/glove)

Neck (µg/glove)

Collar (µg/glove)

5

81.81 ± 19.0

9312 ± 1624

22,413 ± 2907

12

82.12 ± 32.0

6738 ± 1091

15,788 ± 2101


Tetrachlorvinphos

NOTE: Pretreatment value for tetrachlorvinphos was 3.03 ± 1.53 µg/glove.

Table VIII. Tetrachlorvinphos Tee Shirt Residues for the Short-term Studies (concentration ± standard error)

Tetrachlorvinphos

Day 8 9 10 11 12

ng/g Shirt 1,692 ± 657 1,010 ± 435 2,075 ± 1,031 1,026 ± 277 1,625 ± 926

The TCVP metabolite TCMA was quantified over 5 days in adults and children on the second week after placement of the collar. In contrast to chlorpyrifos, the pretreatment baseline levels of TCMA in both adults and children were very low, about 1.7 ng/ml (Table IX). These low baseline residues reflect the fact that TCVP is not a widely used insecticide and therefore the likelihood of TCVP exposure of people is very low. The residues of TCMA in adult urine were 43-104 ng/ml and the residues of TCMA in children’s urine were 164-199 ng/ml in the post-treatment samples. These post-treatment samples are very clearly above the pretreatment values by one or two orders of magnitude, and were significantly different from pretreatment values (P < 0.05). Similar to the urinary metabolites from the chlorpyrifos collar, the TCMA levels in the urine of children were higher than those in adults. While the levels of these transferable residues and urinary metabolites seem very high, especially when compared to those from chlorpyrifos and TCP, TCVP is a very low toxicity insecticide and is probably metabolically detoxified very quickly, leading to the production of TCMA quickly. Therefore the significance of these high residues as related to hazard is unknown, and these numbers should not be construed at this point to be a cause for concern. If there are poor metabolizers in the population with respect to TCVP, these levels of residues might be of greater concern to such a sub-group, but we have no information about this possibility.


171 Table IX. Urinary TCMA Concentrations for the Short-term Studies (concentration ± standard error)


TCMA (ng/mL Urine)

Day 8 9 10 11 12

Adult 56.80 ± 13.91 65.74 ± 24.88 55.49 ± 18.88 43.94 ± 13.66 103.56 ± 37.92

Child 175.04 ± 44.25 164.17 ± 43.91 172.57 ± 60.87 198.89 ± 73.74 161.80 ± 56.79

NOTE: Pretreatment values for TCMA = 1.74 ± 0.95 (combined average for adults and children).

It should be noted that with both products the methods employed to assay the residues in either gloves or tee shirts quantified only the parent insecticide and not any breakdown products. Thus it is possible that TCP could have been available for absorption as well as chlorpyrifos and TCMA as well as TCVP; if this were the case, then contributions of these breakdown products could have been made to the urinary metabolite levels observed. It cannot be determined from the current data whether any absorption with subsequent excretion of the breakdown products was occurring, but is certainly a possibility. However, the levels of parent insecticide observed in gloves and tee shirts indicate that the parent compounds were certainly available for absorption.

Summary and Conclusions Our laboratories have developed protocols for the sampling of transferable residues from the fur of dogs treated with topical flea control products. With the two collars containing organophosphorus insecticides reported upon here, we obtained similar and logical results in the pattern of transferable residues from the fur of the dogs, with the back distant from the collar having the lowest transferable residues and the samples taken over the collar having the highest residues. However, the residues from the TCVP collar were considerably higher than those from the chlorpyrifos collar. These differences in magnitude can probably be attributed to differences in the polymer matrix used for the collar. Because TCVP is of considerably lower acute mammalian toxicity than chlorpyrifos, it is probably not of concern to the pet or to the people that the TCVP migrates out of its collar faster than chlorpyrifos migrates out of its collar. These postulated rates of migration out of the collar are probably the reason that the TCVP collar is only recommended for 4 months while the chlorpyrifos collar is recommended for 11 months, despite the fact that the TCVP collar has a higher percentage of active ingredient than does the chlorpyrifos collar. We had hoped that the tee shirt data might be a good surrogate for exposure, but there were no useful correlations between the tee shirt residues and the biomonitoring data. The biomonitoring data showed consistently higher



172 residues of urinary metabolites in children compared to adults with both insecticides. In addition to the data shown here, the urinary metabolite data were also calculated corrected for urinary creatinine values, and these calculations did not change appreciably the relationship between adult and children’s urine. It appears that little, if any, additional exposure to chlorpyrifos occurs because of the use of this collar, although the data are still somewhat equivocal with respect to children’s exposure. However, when comparing our metabolite data to the published NHANES data, the ranges in the geometric means we observed for each age group (children - 8.43 to 19.71 ng/mL urine; adults - 4.79 to 8.58 ng/mL urine) were similar to those in the 95th and 90th percentile of the NHANES data, respectively (8). In our studies, we feel that some exposure to TCVP occurred because of the use of the collar. This is further supported by comparing the ranges in the geometric means for our data to the published NHANES data. The range of urinary TCMA in children was 30.84 to 55.25 ng/mL urine, and the range in adults was 12.90 to 20.57 ng/mL urine. Both of the ranges were well above the 95th percentile rankings reported in the NHANES data (8). However, because of the low toxicity of TCVP, the significance of this exposure in the risk assessment cannot be concluded at this time.

Acknowledgements This research was supported by grants from the U.S. Environmental Protection Agency’s Science to Achieve Results (STAR) grant program (Grant Numbers R825170 and R828017). Although the research described herein has been funded wholly or in part by the U. S. Environmental Protection Agency STAR program, it has not been subjected to any governmental review and therefore does not reflect the views of the agency. No official endorsements should be inferred.

References 1. 2. 3. 4. 5. 6.

American Veterinary Medical Association. US Pet Ownership and Demographics Sourcebook; Center for Information Management, American Veterinary Association: Schaumburg, IL. 1997. Ecobichon, D.J. In Casarett and Doull’s Toxicology, 5th ed.; Klaassen, C.D., Ed.; McGraw-Hill: New York, NY, 1996, pp. 643-698. Boone, J.S., J.W. Tyler, and J.E. Chambers. Environ. Hlth. Perspect. 2001, 109, 1109-1114. Boone, J.S.; Tyler, J.W.; Davis, M.K.; Chambers, J.E. Tox. Mechanisms Methods. 2006, 16, 275-280. Chambers, J.E.; Boone, J.S.; Davis, M.K.; Moran, J.E.; Tyler, J.W. J. Expo. Sci. Environ. Epidemiol. 2007, 17, 656-666. Davis, M.K.; Boone, J.S.; Moran, J.E.; Tyler, J.W.; Chambers, J.E. J. Expo. Sci. Environ. Epidemiol. 2008, URL http://www.nature.com/doifinder/10.1038/sj.jes.7500647.


173 7. 8.


9.

USEPA. Organophosphate Pesticides in Food – Primer on Reassessment of Residue Limits; United States Environmental Protection Agency: Washington, D.C., 1999. National Center for Environmental Health. Third National Report on Human Exposure to Environmental Chemicals; U.S. Department of Health and Human Services, Centers for Disease Control and Prevention: Atlanta, GA. 2005, NCEH Pub. No. 05-0570. USEPA. TCVP Registration Eligibility Decision; United States Environmental Protection Agency: Washington, D.C., 1995.


bk-2009-1015.ch012

Recommend Documents